if (!require(titanic)) install.packages(“titanic”) library(dplyr) library(ggplot2) library(titanic)

titanic_df <- titanic::titanic_train

titanic_data <- titanic_df %>% select(Survived, Pclass, Sex, Age, Fare) %>% na.omit()

titanic_data\(Sex <- as.numeric(factor(titanic_data\)Sex, levels = c(“male”, “female”))) titanic_data\(Survived <- as.factor(titanic_data\)Survived)

head(titanic_data) # Xây dựng mô hình logistic logistic_model <- glm(Survived ~ Pclass + Sex + Age + Fare, data = titanic_data, family = binomial)

tóm tắt mô hình

summary(logistic_model) library(caret) pred <- predict(logistic_model, titanic_data, type = “response”) pred_class <- ifelse(pred > 0.5, 1, 0)

Chuyển về dạng factor

pred_class <- as.factor(pred_class) titanic_data\(Survived <- as.factor(titanic_data\)Survived)

Tính F1-score

confusionMatrix(pred_class, titanic_data\(Survived, mode = "everything")\)byClass[“F1”]